The Apriori Algorithm is an unsupervised learning technique for producing associative rules. This talk will explain the algorithm's implementation, explore how effective it can be when applied to big data, discuss how we use it at DataScience to do market basket analysis, and demonstrate some novel use cases involving the million song database, recipes, and other applications involving open data.
PayPal prvoides an online transfer money network. Each payment flow connects senders and receivers into a giant network where each sender/receiver is a node and each transaction is an edge. Traditionally, the risk score of a transaction is computed based on the characteristics of the involved sender/receiver/transaction. In this talk, we will describe a novel network inference approach to calculate transaction risk score that also includes the risk profile of neighboring senders and receivers using Apache Giraph. The approach reveals additional risk insights not possible with the traditional method. We leverage Hadoop to support a graph computation involving hundreds of millions of nodes and edges.
Tugdual Grall - Real World Use Cases: Hadoop and NoSQL in ProductionCodemotion
What’s important about a technology is what you can use it to do. I’ve looked at what a number of groups are doing with Apache Hadoop and NoSQL in production, and I will relay what worked well for them and what did not. Drawing from real world use cases, I show how people who understand these new approaches can employ them well in conjunction with traditional approaches and existing applications. Thread Detection, Datawarehouse optimization, Marketing Efficiency, Biometric Database are some examples exposed during this presentation.
HBaseCon 2015: Warcbase - Scaling 'Out' and 'Down' HBase for Web ArchivingHBaseCon
Web archiving initiatives around the world capture ephemeral web content to preserve our collective digital memory. However, that requires scalable, responsive tools that support exploration and discovery of captured content. Here you'll learn about why Warcbase, an open-source platform for managing web archives built on HBase, is one such tool. It provides a flexible data model for storing and managing raw content as well as metadata and extracted knowledge, tightly integrates with Hadoop for analytics and data processing, and relies on HBase for storage infrastructure.
Big Data Day LA 2016/ Big Data Track - Twitter Heron @ Scale - Karthik Ramasa...Data Con LA
Twitter generates billions and billions of events per day. Analyzing these events in real time presents a massive challenge. Twitter designed and deployed a new streaming system called Heron. Heron has been in production nearly 2 years and is widely used by several teams for diverse use cases. This talk looks at Twitter's operating experiences and challenges of running Heron at scale and the approaches taken to solve those challenges.
Rainbird: Realtime Analytics at Twitter (Strata 2011)Kevin Weil
Introducing Rainbird, Twitter's high volume distributed counting service for realtime analytics, built on Cassandra. This presentation looks at the motivation, design, and uses of Rainbird across Twitter.
Watching Pigs Fly with the Netflix Hadoop Toolkit (Hadoop Summit 2013)Jeff Magnusson
Overview of the data platform as a service architecture at Netflix. We examine the tools and services built around the Netflix Hadoop platform that are designed to make access to big data at Netflix easy, efficient, and self-service for our users.
From the perspective of a user of the platform, we walk through how various services in the architecture can be used to build a recommendation engine. Sting, a tool for fast in memory aggregation and data visualization, and Lipstick, our workflow visualization and monitoring tool for Apache Pig, are discussed in depth. Lipstick is now part of Netflix OSS - clone it on github, or learn more from our techblog post: http://techblog.netflix.com/2013/06/introducing-lipstick-on-apache-pig.html.
PayPal prvoides an online transfer money network. Each payment flow connects senders and receivers into a giant network where each sender/receiver is a node and each transaction is an edge. Traditionally, the risk score of a transaction is computed based on the characteristics of the involved sender/receiver/transaction. In this talk, we will describe a novel network inference approach to calculate transaction risk score that also includes the risk profile of neighboring senders and receivers using Apache Giraph. The approach reveals additional risk insights not possible with the traditional method. We leverage Hadoop to support a graph computation involving hundreds of millions of nodes and edges.
Tugdual Grall - Real World Use Cases: Hadoop and NoSQL in ProductionCodemotion
What’s important about a technology is what you can use it to do. I’ve looked at what a number of groups are doing with Apache Hadoop and NoSQL in production, and I will relay what worked well for them and what did not. Drawing from real world use cases, I show how people who understand these new approaches can employ them well in conjunction with traditional approaches and existing applications. Thread Detection, Datawarehouse optimization, Marketing Efficiency, Biometric Database are some examples exposed during this presentation.
HBaseCon 2015: Warcbase - Scaling 'Out' and 'Down' HBase for Web ArchivingHBaseCon
Web archiving initiatives around the world capture ephemeral web content to preserve our collective digital memory. However, that requires scalable, responsive tools that support exploration and discovery of captured content. Here you'll learn about why Warcbase, an open-source platform for managing web archives built on HBase, is one such tool. It provides a flexible data model for storing and managing raw content as well as metadata and extracted knowledge, tightly integrates with Hadoop for analytics and data processing, and relies on HBase for storage infrastructure.
Big Data Day LA 2016/ Big Data Track - Twitter Heron @ Scale - Karthik Ramasa...Data Con LA
Twitter generates billions and billions of events per day. Analyzing these events in real time presents a massive challenge. Twitter designed and deployed a new streaming system called Heron. Heron has been in production nearly 2 years and is widely used by several teams for diverse use cases. This talk looks at Twitter's operating experiences and challenges of running Heron at scale and the approaches taken to solve those challenges.
Rainbird: Realtime Analytics at Twitter (Strata 2011)Kevin Weil
Introducing Rainbird, Twitter's high volume distributed counting service for realtime analytics, built on Cassandra. This presentation looks at the motivation, design, and uses of Rainbird across Twitter.
Watching Pigs Fly with the Netflix Hadoop Toolkit (Hadoop Summit 2013)Jeff Magnusson
Overview of the data platform as a service architecture at Netflix. We examine the tools and services built around the Netflix Hadoop platform that are designed to make access to big data at Netflix easy, efficient, and self-service for our users.
From the perspective of a user of the platform, we walk through how various services in the architecture can be used to build a recommendation engine. Sting, a tool for fast in memory aggregation and data visualization, and Lipstick, our workflow visualization and monitoring tool for Apache Pig, are discussed in depth. Lipstick is now part of Netflix OSS - clone it on github, or learn more from our techblog post: http://techblog.netflix.com/2013/06/introducing-lipstick-on-apache-pig.html.
Slidedeck from the InfoFarm Real Time Big Data Seminar. Main Topics are: Apache Kafka, Apache Spark, Apache Storm and integration and visualisations with Elasticsearch and Kibana.
Pablo Musa - Managing your Black Friday Logs - Codemotion Amsterdam 2019Codemotion
Monitoring an entire application is not a simple task, but with the right tools it is not a hard task either. However, events like Black Friday can push your application to the limit, and even cause crashes. As the system is stressed, it generates a lot more logs, which may crash the monitoring system as well. In this talk I will walk through the best practices when using the Elastic Stack to centralize and monitor your logs. I will also share some tricks to help you with the huge increase of traffic typical in Black Fridays.
Democratizing Machine Learning: Perspective from a scikit-learn CreatorDatabricks
<p>Once an obscure branch of applied mathematics, machine learning is now the darling of tech. I will talk about lessons learned democratizing machine learning. How libraries like scikit-learn were designed to empower users: simplifying but avoiding ambiguous behaviors. How the Python data ecosystem was built from scientific computing tools: the importance of good numerics. How some machine-learning patterns easily provide value to real-world situations. I will also discuss remain challenges to address and the progresses that we are making. Scaling up brings different bottlenecks to numerics. Integrating data in the statistical models, a hurdle to data-science practice requires to rethink data cleaning pipelines.</p><p>This talk will drawn from my experience as a scikit-learn developer, but also as a researcher in machine learning and applications.</p>
Blue Pill/Red Pill: The Matrix of Thousands of Data StreamsDatabricks
Designing a streaming application which has to process data from 1 or 2 streams is easy. Any streaming framework which provides scalability, high-throughput, and fault-tolerance would work. But when the number of streams start growing in order 100s or 1000s, managing them can be daunting. How would you share resources among 1000s of streams with all of them running 24×7? Manage their state, Apply advanced streaming operations, Add/Delete streams without restarting? This talk explains common scenarios & shows techniques that can handle thousands of streams using Spark Structured Streaming.
Resilience: the key requirement of a [big] [data] architecture - StampedeCon...StampedeCon
From the StampedeCon 2015 Big Data Conference: There is an adage, “If you fail to plan, you plan to fail” . When developing systems the adage can be taken a step further, “If you fail to plan FOR FAILURE, you plan to fail”. At Huffington post data moves between a number of systems to provide statistics for our technical, business, and editorial teams. Due to the mission-critical nature of our data, considerable effort is spent building resiliency into processes.
This talk will focus on designing for failure. Some material will focus understanding the traits of specific distributed systems such as message queues or NoSQL databases and what are the consequences for different types of failures. While other parts of the presentation will focus on how systems and software can be designed to make re-processing batch data simple, or how to determine what failure mode semantics are important for a real time event processing system.
Janus graph lookingbackwardreachingforwardDemai Ni
JanusGraph: Looking Backward and Reaching Forward - by Jason Plurad (@pluradj):
The JanusGraph project started at the Linux Foundation earlier this year, but it is not the new kid on the block. We'll start with a look at the origins and evolution of this open source graph database through the lens of a few IBM graph use cases. We'll discuss the new features in latest release of JanusGraph, and then take a look at future directions to explore together with the open community.
AI on Spark for Malware Analysis and Anomalous Threat DetectionDatabricks
At Avast, we believe everyone has the right to be safe. We are dedicated to creating a world that provides safety and privacy for all, not matter where you are, who you are, or how you connect. With over 1.5 billion attacks stopped and 30 million new executable files monthly, big data pipelines are crucial for the security of our customers. At Avast we are leveraging Apache Spark machine learning libraries and TensorflowOnSpark for a variety of tasks ranging from marketing and advertisement, through network security to malware detection. This talk will cover our main cybersecurity usecases of Spark. After describing our cluster environment we will first demonstrate anomaly detection on time series of threats. Having thousands of types of attacks and malware, AI helps human analysts select and focus on most urgent or dire threats. We will walk through our setup for distributed training of deep neural networks with Tensorflow to deploying and monitoring of a streaming anomaly detection application with trained model. Next we will show how we use Spark for analysis and clustering of malicious files and large scale experimentation to automatically process and handle changes in malware. In the end, we will give comparison to other tools we used for solving those problems.
High-Performance Advanced Analytics with Spark-AlchemyDatabricks
Pre-aggregation is a powerful analytics technique as long as the measures being computed are reaggregable. Counts reaggregate with SUM, minimums with MIN, maximums with MAX, etc. The odd one out is distinct counts, which are not reaggregable.
Traditionally, the non-reaggregability of distinct counts leads to an implicit restriction: whichever system computes distinct counts has to have access to the most granular data and touch every row at query time. Because of this, in typical analytics architectures, where fast query response times are required, raw data has to be duplicated between Spark and another system such as an RDBMS. This talk is for everyone who computes or consumes distinct counts and for everyone who doesn’t understand the magical power of HyperLogLog (HLL) sketches.
We will break through the limits of traditional analytics architectures using the advanced HLL functionality and cross-system interoperability of the spark-alchemy open-source library, whose capabilities go beyond what is possible with OSS Spark, Redshift or even BigQuery. We will uncover patterns for 1000x gains in analytic query performance without data duplication and with significantly less capacity.
We will explore real-world use cases from Swoop’s petabyte-scale systems, improve data privacy when running analytics over sensitive data, and even see how a real-time analytics frontend running in a browser can be provisioned with data directly from Spark.
Real Time Processing Using Twitter Heron by Karthik RamasamyData Con LA
Abstract:- Today's enterprises are not only producing data in high volume but also at high velocity. With velocity comes the need to process the data in real time. To meet the real time needs, we developed and deployed Heron, the next generation streaming engine at Twitter. Heron processes billions and billions of events per day at Twitter and has been in production for nearly 3 years. Heron provides unparalleled performance at large scale and has been successfully meeting Twitter's strict performance requirements for various streaming and iOT applications. Heron is a open source project with several major contributors from various institutions. As the project, we identified and implemented several optimizations that improved throughput by additional 5x and further reduce latency by 50-60%. In this talk, we will describe Heron in detail, how the detailed profiling indicated the performance bottleneck areas such as multiple serializations/deserialization and immutable data structures. After mitigating these costs, we were able to show much higher throughput and latencies as low as 12ms.
R + Storm Moneyball - Realtime Advanced Statistics - Hadoop Summit - San JoseAllen Day, PhD
Architecting R into the Storm Application Development Process
~~~~~
The business need for real-time analytics at large scale has focused attention on the use of Apache Storm, but an approach that is sometimes overlooked is the use of Storm and R together. This novel combination of real-time processing with Storm and the practical but powerful statistical analysis offered by R substantially extends the usefulness of Storm as a solution to a variety of business critical problems. By architecting R into the Storm application development process, Storm developers can be much more effective. The aim of this design is not necessarily to deploy faster code but rather to deploy code faster. Just a few lines of R code can be used in place of lengthy Storm code for the purpose of early exploration – you can easily evaluate alternative approaches and quickly make a working prototype.
In this presentation, Allen will build a bridge from basic real-time business goals to the technical design of solutions. We will take an example of a real-world use case, compose an implementation of the use case as Storm components (spouts, bolts, etc.) and highlight how R can be an effective tool in prototyping a solution.
Advanced Analytics for Any Data at Real-Time Speeddanpotterdwch
The kenyote presentation from Predictive Analytics World entitled "Advanced Analytics for Any Data at Real-Time Speed" Dan Potter, CMO from Datawatch, presents a new approach to prepare, incorporate, enrich and visualize streaming data for advanced visual analysis is essential for making timelier, high-impact business decisions in tough competitive markets.
JanusGraph: What's Next, Project Status Update. Presented at Open Source Graph Technologies NYC Meetup on August 24, 2017. https://www.meetup.com/graphs/events/241136321/
Virdata: lessons learned from the Internet of Things and M2M Cloud Services @...Nathan Bijnens
Presentation I gave at the IBM Big Data Developers meetup group in San Jose, CA.
There is also a video available of this talk at:
https://www.youtube.com/watch?v=TSt49yPBmW0&t=7m59s
Speaker: Philippe Mizrahi - Associate Product Manager - Lyft
Abstract: Philippe Mizrahi works on Lyft’s data discovery and metadata engine, Amundsen. With the help of a Neo4j graph database, Amundsen has improved Lyft’s data discovery by reducing time to discover data by 10x.
During this session, Philippe will dive deep into Amundsen’s use cases, impact, and architecture, which effectively combines a comprehensive knowledge graph based upon Neo4j, centralized metadata and other search ranking optimizations to discover data quickly.
Concepts, use cases and principles to build big data systems (1)Trieu Nguyen
1) Introduction to the key Big Data concepts
1.1 The Origins of Big Data
1.2 What is Big Data ?
1.3 Why is Big Data So Important ?
1.4 How Is Big Data Used In Practice ?
2) Introduction to the key principles of Big Data Systems
2.1 How to design Data Pipeline in 6 steps
2.2 Using Lambda Architecture for big data processing
3) Practical case study : Chat bot with Video Recommendation Engine
4) FAQ for student
Slidedeck from the InfoFarm Real Time Big Data Seminar. Main Topics are: Apache Kafka, Apache Spark, Apache Storm and integration and visualisations with Elasticsearch and Kibana.
Pablo Musa - Managing your Black Friday Logs - Codemotion Amsterdam 2019Codemotion
Monitoring an entire application is not a simple task, but with the right tools it is not a hard task either. However, events like Black Friday can push your application to the limit, and even cause crashes. As the system is stressed, it generates a lot more logs, which may crash the monitoring system as well. In this talk I will walk through the best practices when using the Elastic Stack to centralize and monitor your logs. I will also share some tricks to help you with the huge increase of traffic typical in Black Fridays.
Democratizing Machine Learning: Perspective from a scikit-learn CreatorDatabricks
<p>Once an obscure branch of applied mathematics, machine learning is now the darling of tech. I will talk about lessons learned democratizing machine learning. How libraries like scikit-learn were designed to empower users: simplifying but avoiding ambiguous behaviors. How the Python data ecosystem was built from scientific computing tools: the importance of good numerics. How some machine-learning patterns easily provide value to real-world situations. I will also discuss remain challenges to address and the progresses that we are making. Scaling up brings different bottlenecks to numerics. Integrating data in the statistical models, a hurdle to data-science practice requires to rethink data cleaning pipelines.</p><p>This talk will drawn from my experience as a scikit-learn developer, but also as a researcher in machine learning and applications.</p>
Blue Pill/Red Pill: The Matrix of Thousands of Data StreamsDatabricks
Designing a streaming application which has to process data from 1 or 2 streams is easy. Any streaming framework which provides scalability, high-throughput, and fault-tolerance would work. But when the number of streams start growing in order 100s or 1000s, managing them can be daunting. How would you share resources among 1000s of streams with all of them running 24×7? Manage their state, Apply advanced streaming operations, Add/Delete streams without restarting? This talk explains common scenarios & shows techniques that can handle thousands of streams using Spark Structured Streaming.
Resilience: the key requirement of a [big] [data] architecture - StampedeCon...StampedeCon
From the StampedeCon 2015 Big Data Conference: There is an adage, “If you fail to plan, you plan to fail” . When developing systems the adage can be taken a step further, “If you fail to plan FOR FAILURE, you plan to fail”. At Huffington post data moves between a number of systems to provide statistics for our technical, business, and editorial teams. Due to the mission-critical nature of our data, considerable effort is spent building resiliency into processes.
This talk will focus on designing for failure. Some material will focus understanding the traits of specific distributed systems such as message queues or NoSQL databases and what are the consequences for different types of failures. While other parts of the presentation will focus on how systems and software can be designed to make re-processing batch data simple, or how to determine what failure mode semantics are important for a real time event processing system.
Janus graph lookingbackwardreachingforwardDemai Ni
JanusGraph: Looking Backward and Reaching Forward - by Jason Plurad (@pluradj):
The JanusGraph project started at the Linux Foundation earlier this year, but it is not the new kid on the block. We'll start with a look at the origins and evolution of this open source graph database through the lens of a few IBM graph use cases. We'll discuss the new features in latest release of JanusGraph, and then take a look at future directions to explore together with the open community.
AI on Spark for Malware Analysis and Anomalous Threat DetectionDatabricks
At Avast, we believe everyone has the right to be safe. We are dedicated to creating a world that provides safety and privacy for all, not matter where you are, who you are, or how you connect. With over 1.5 billion attacks stopped and 30 million new executable files monthly, big data pipelines are crucial for the security of our customers. At Avast we are leveraging Apache Spark machine learning libraries and TensorflowOnSpark for a variety of tasks ranging from marketing and advertisement, through network security to malware detection. This talk will cover our main cybersecurity usecases of Spark. After describing our cluster environment we will first demonstrate anomaly detection on time series of threats. Having thousands of types of attacks and malware, AI helps human analysts select and focus on most urgent or dire threats. We will walk through our setup for distributed training of deep neural networks with Tensorflow to deploying and monitoring of a streaming anomaly detection application with trained model. Next we will show how we use Spark for analysis and clustering of malicious files and large scale experimentation to automatically process and handle changes in malware. In the end, we will give comparison to other tools we used for solving those problems.
High-Performance Advanced Analytics with Spark-AlchemyDatabricks
Pre-aggregation is a powerful analytics technique as long as the measures being computed are reaggregable. Counts reaggregate with SUM, minimums with MIN, maximums with MAX, etc. The odd one out is distinct counts, which are not reaggregable.
Traditionally, the non-reaggregability of distinct counts leads to an implicit restriction: whichever system computes distinct counts has to have access to the most granular data and touch every row at query time. Because of this, in typical analytics architectures, where fast query response times are required, raw data has to be duplicated between Spark and another system such as an RDBMS. This talk is for everyone who computes or consumes distinct counts and for everyone who doesn’t understand the magical power of HyperLogLog (HLL) sketches.
We will break through the limits of traditional analytics architectures using the advanced HLL functionality and cross-system interoperability of the spark-alchemy open-source library, whose capabilities go beyond what is possible with OSS Spark, Redshift or even BigQuery. We will uncover patterns for 1000x gains in analytic query performance without data duplication and with significantly less capacity.
We will explore real-world use cases from Swoop’s petabyte-scale systems, improve data privacy when running analytics over sensitive data, and even see how a real-time analytics frontend running in a browser can be provisioned with data directly from Spark.
Real Time Processing Using Twitter Heron by Karthik RamasamyData Con LA
Abstract:- Today's enterprises are not only producing data in high volume but also at high velocity. With velocity comes the need to process the data in real time. To meet the real time needs, we developed and deployed Heron, the next generation streaming engine at Twitter. Heron processes billions and billions of events per day at Twitter and has been in production for nearly 3 years. Heron provides unparalleled performance at large scale and has been successfully meeting Twitter's strict performance requirements for various streaming and iOT applications. Heron is a open source project with several major contributors from various institutions. As the project, we identified and implemented several optimizations that improved throughput by additional 5x and further reduce latency by 50-60%. In this talk, we will describe Heron in detail, how the detailed profiling indicated the performance bottleneck areas such as multiple serializations/deserialization and immutable data structures. After mitigating these costs, we were able to show much higher throughput and latencies as low as 12ms.
R + Storm Moneyball - Realtime Advanced Statistics - Hadoop Summit - San JoseAllen Day, PhD
Architecting R into the Storm Application Development Process
~~~~~
The business need for real-time analytics at large scale has focused attention on the use of Apache Storm, but an approach that is sometimes overlooked is the use of Storm and R together. This novel combination of real-time processing with Storm and the practical but powerful statistical analysis offered by R substantially extends the usefulness of Storm as a solution to a variety of business critical problems. By architecting R into the Storm application development process, Storm developers can be much more effective. The aim of this design is not necessarily to deploy faster code but rather to deploy code faster. Just a few lines of R code can be used in place of lengthy Storm code for the purpose of early exploration – you can easily evaluate alternative approaches and quickly make a working prototype.
In this presentation, Allen will build a bridge from basic real-time business goals to the technical design of solutions. We will take an example of a real-world use case, compose an implementation of the use case as Storm components (spouts, bolts, etc.) and highlight how R can be an effective tool in prototyping a solution.
Advanced Analytics for Any Data at Real-Time Speeddanpotterdwch
The kenyote presentation from Predictive Analytics World entitled "Advanced Analytics for Any Data at Real-Time Speed" Dan Potter, CMO from Datawatch, presents a new approach to prepare, incorporate, enrich and visualize streaming data for advanced visual analysis is essential for making timelier, high-impact business decisions in tough competitive markets.
JanusGraph: What's Next, Project Status Update. Presented at Open Source Graph Technologies NYC Meetup on August 24, 2017. https://www.meetup.com/graphs/events/241136321/
Virdata: lessons learned from the Internet of Things and M2M Cloud Services @...Nathan Bijnens
Presentation I gave at the IBM Big Data Developers meetup group in San Jose, CA.
There is also a video available of this talk at:
https://www.youtube.com/watch?v=TSt49yPBmW0&t=7m59s
Speaker: Philippe Mizrahi - Associate Product Manager - Lyft
Abstract: Philippe Mizrahi works on Lyft’s data discovery and metadata engine, Amundsen. With the help of a Neo4j graph database, Amundsen has improved Lyft’s data discovery by reducing time to discover data by 10x.
During this session, Philippe will dive deep into Amundsen’s use cases, impact, and architecture, which effectively combines a comprehensive knowledge graph based upon Neo4j, centralized metadata and other search ranking optimizations to discover data quickly.
Concepts, use cases and principles to build big data systems (1)Trieu Nguyen
1) Introduction to the key Big Data concepts
1.1 The Origins of Big Data
1.2 What is Big Data ?
1.3 Why is Big Data So Important ?
1.4 How Is Big Data Used In Practice ?
2) Introduction to the key principles of Big Data Systems
2.1 How to design Data Pipeline in 6 steps
2.2 Using Lambda Architecture for big data processing
3) Practical case study : Chat bot with Video Recommendation Engine
4) FAQ for student
Rethink Analytics with an Enterprise Data HubCloudera, Inc.
Have you run into one or more of the following barriers or limitations with your existing data warehousing architecture:
> Increasingly high data storage and/or processing costs?
> Silos of data sources?
> Complexity of management and security?
> Lack of analytics agility?
An introductory but highly practical talk on starting a Data Science career and life. It touches upon all the main aspects of the path towards becoming a Data scientist, also seen through a personal development perspective. Moreover, we talk about the role that a data scientist ultimately fulfills - as an individual or as a team - in the technology innovation life cycle and the product life-cycle.
Using BigBench to compare Hive and Spark (Long version)Nicolas Poggi
BigBench is the brand new standard (TPCx-BB) for benchmarking and testing Big Data systems. The BigBench specification describes several application use cases combining the need for SQL queries, Map/Reduce, user code (UDF), Machine Learning, and even streaming. From the available implementation, we can test the different framework combinations such as Hadoop+Hive (with Mahout) and Spark (SparkSQL+MLlib) in their different versions and configurations, helping us to spot problems and possible optimizations of our data stacks.
This talk first introduces BigBench and how problems can it solve. Then, presents both Hive and Spark benchmark results with their respective 1 and 2 versions under distinct configurations including Tez, Mahout, MLlib. Experiments are run on Cloud and On-Prem clusters of different numbers of nodes and data scales, taking into account interactive and batch usage. Results are further classified by use cases, showing where each platform shines (or doesn't), and why, based on performance metrics and logfile analysis. The talk concludes with the main findings, the scalability, and limits of each framework.
Originally presented at: https://dataworkssummit.com/munich-2017/sessions/using-bigbench-to-compare-hive-and-spark-versions-and-features/
Talk on Data Discovery and Metadata by Mark Grover from July 2019.
Goes into detail of the problem, build/buy/adopt analysis and Lyft's solution - Amundsen, along with thoughts on the future.
Data Con LA 2022 - Using Google trends data to build product recommendationsData Con LA
Mike Limcaco, Analytics Specialist / Customer Engineer at Google
Measure trends in a particular topic or search term across Google Search across the US down to the city-level. Integrate these data signals into analytic pipelines to drive product, retail, media (video, audio, digital content) recommendations tailored to your audience segment. We'll discuss how Google unique datasets can be used with Google Cloud smart analytic services to process, enrich and surface the most relevant product or content that matches the ever-changing interests of your local customer segment.
Melinda Thielbar, Data Science Practice Lead and Director of Data Science at Fidelity Investments
From corporations to governments to private individuals, most of the AI community has recognized the growing need to incorporate ethics into the development and maintenance of AI models. Much of the current discussion, though, is meant for leaders and managers. This talk is directed to data scientists, data engineers, ML Ops specialists, and anyone else who is responsible for the hands-on, day-to-day of work building, productionalizing, and maintaining AI models. We'll give a short overview of the business case for why technical AI expertise is critical to developing an AI Ethics strategy. Then we'll discuss the technical problems that cause AI models to behave unethically, how to detect problems at all phases of model development, and the tools and techniques that are available to support technical teams in Ethical AI development.
Data Con LA 2022 - Improving disaster response with machine learningData Con LA
Antje Barth, Principal Developer Advocate, AI/ML at AWS & Chris Fregly, Principal Engineer, AI & ML at AWS
The frequency and severity of natural disasters are increasing. In response, governments, businesses, nonprofits, and international organizations are placing more emphasis on disaster preparedness and response. Many organizations are accelerating their efforts to make their data publicly available for others to use. Repositories such as the Registry of Open Data on AWS and Humanitarian Data Exchange contain troves of data available for use by developers, data scientists, and machine learning practitioners. In this session, see how a community of developers came together though the AWS Disaster Response hackathon to build models to support natural disaster preparedness and response.
Data Con LA 2022 - What's new with MongoDB 6.0 and AtlasData Con LA
Sig Narvaez, Executive Solution Architect at MongoDB
MongoDB is now a Developer Data Platform. Come learn what�s new in the 6.0 release and Atlas following all the recent announcements made at MongoDB World 2022. Topics will include
- Atlas Search which combines 3 systems into one (database, search engine, and sync mechanisms) letting you focus on your product's differentiation.
- Atlas Data Federation to seamlessly query, transform, and aggregate data from one or more MongoDB Atlas databases, Atlas Data Lake and AWS S3 buckets
- Queryable Encryption lets you run expressive queries on fully randomized encrypted data to meet the most stringent security requirements
- Relational Migrator which analyzes your existing relational schemas and helps you design a new MongoDB schema.
- And more!
Data Con LA 2022 - Real world consumer segmentationData Con LA
Jaysen Gillespie, Head of Analytics and Data Science at RTB House
1. Shopkick has over 30M downloads, but the userbase is very heterogeneous. Anecdotal evidence indicated a wide variety of users for whom the app holds long-term appeal.
2. Marketing and other teams challenged Analytics to get beyond basic summary statistics and develop a holistic segmentation of the userbase.
3. Shopkick's data science team used SQL and python to gather data, clean data, and then perform a data-driven segmentation using a k-means algorithm.
4. Interpreting the results is more work -- and more fun -- than running the algo itself. We'll discuss how we transform from ""segment 1"", ""segment 2"", etc. to something that non-analytics users (Marketing, Operations, etc.) could actually benefit from.
5. So what? How did team across Shopkick change their approach given what Analytics had discovered.
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...Data Con LA
Ravi Pillala, Chief Data Architect & Distinguished Engineer at Intuit
TurboTax is one of the well known consumer software brand which at its peak serves 385K+ concurrent users. In this session, We start with looking at how user behavioral data & tax domain events are captured in real time using the event bus and analyzed to drive real time personalization with various TurboTax data pipelines. We will also look at solutions performing analytics which make use of these events, with the help of Kafka, Apache Flink, Apache Beam, Spark, Amazon S3, Amazon EMR, Redshift, Athena and Amazon lambda functions. Finally, we look at how SageMaker is used to create the TurboTax model to predict if a customer is at risk or needs help.
Data Con LA 2022 - Moving Data at Scale to AWSData Con LA
George Mansoor, Chief Information Systems Officer at California State University
Overview of the CSU Data Architecture on moving on-prem ERP data to the AWS Cloud at scale using Delphix for Data Replication/Virtualization and AWS Data Migration Service (DMS) for data extracts
Data Con LA 2022 - Collaborative Data Exploration using Conversational AIData Con LA
Anand Ranganathan, Chief AI Officer at Unscrambl
Conversational AI is getting more and more widely used for customer support and employee support use-cases. In this session, I'm going to talk about how it can be extended for data analysis and data science use-cases ... i.e., how users can interact with a bot to ask analytical questions on data in relational databases.
This allows users to explore complex datasets using a combination of text and voice questions, in natural language, and then get back results in a combination of natural language and visualizations. Furthermore, it allows collaborative exploration of data by a group of users in a channel in platforms like Microsoft Teams, Slack or Google Chat.
For example, a group of users in a channel can ask questions to a bot in plain English like ""How many cases of Covid were there in the last 2 months by state and gender"" or ""Why did the number of deaths from Covid increase in May 2022"", and jointly look at the results that come back. This facilitates data awareness, data-driven collaboration and joint decision making among teams in enterprises and outside.
In this talk, I'll describe how we can bring together various features including natural-language understanding, NL-to-SQL translation, dialog management, data story-telling, semantic modeling of data and augmented analytics to facilitate collaborate exploration of data using conversational AI.
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...Data Con LA
Anil Inamdar, VP & Head of Data Solutions at Instaclustr
The most modernized enterprises utilize polyglot architecture, applying the best-suited database technologies to each of their organization's particular use cases. To successfully implement such an architecture, though, you need a thorough knowledge of the expansive NoSQL data technologies now available.
Attendees of this Data Con LA presentation will come away with:
-- A solid understanding of the decision-making process that should go into vetting NoSQL technologies and how to plan out their data modernization initiatives and migrations.
-- They will learn the types of functionality that best match the strengths of NoSQL key-value stores, graph databases, columnar databases, document-type databases, time-series databases, and more.
-- Attendees will also understand how to navigate database technology licensing concerns, and to recognize the types of vendors they'll encounter across the NoSQL ecosystem. This includes sniffing out open-core vendors that may advertise as “open source,"" but are driven by a business model that hinges on achieving proprietary lock-in.
-- Attendees will also learn to determine if vendors offer open-code solutions that apply restrictive licensing, or if they support true open source technologies like Hadoop, Cassandra, Kafka, OpenSearch, Redis, Spark, and many more that offer total portability and true freedom of use.
Data Con LA 2022 - Intro to Data ScienceData Con LA
Zia Khan, Computer Systems Analyst and Data Scientist at LearningFuze
Data Science tutorial is designed for people who are new to Data Science. This is a beginner level session so no prior coding or technical knowledge is required. Just bring your laptop with WiFi capability. The session starts with a review of what is data science, the amount of data we generate and how companies are using that data to get insight. We will pick a business use case, define the data science process, followed by hands-on lab using python and Jupyter notebook. During the hands-on portion we will work with pandas, numpy, matplotlib and sklearn modules and use a machine learning algorithm to approach the business use case.
Data Con LA 2022 - How are NFTs and DeFi Changing EntertainmentData Con LA
Mariana Danilovic, Managing Director at Infiom, LLC
We will address:
(1) Community creation and engagement using tokens and NFTs
(2) Organization of DAO structures and ways to incentivize Web3 communities
(3) DeFi business models applied to Web3 ventures
(4) Why Metaverse matters for new entertainment and community engagement models.
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...Data Con LA
Curtis ODell, Global Director Data Integrity at Tricentis
Join me to learn about a new end-to-end data testing approach designed for modern data pipelines that fills dangerous gaps left by traditional data management tools—one designed to handle structured and unstructured data from any source. You'll hear how you can use unique automation technology to reach up to 90 percent test coverage rates and deliver trustworthy analytical and operational data at scale. Several real world use cases from major banks/finance, insurance, health analytics, and Snowflake examples will be presented.
Key Learning Objective
1. Data journeys are complex and you have to ensure integrity of the data end to end across this journey from source to end reporting for compliance
2. Data Management tools do not test data, they profile and monitor at best, and leave serious gaps in your data testing coverage
3. Automation with integration to DevOps and DataOps' CI/CD processes are key to solving this.
4. How this approach has impact in your vertical
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...Data Con LA
Arif Ansari, Professor at University of Southern California
Super Bowl Ad cost $7 million and each year a few Super Bowl ads go viral. The traditional A/B testing does not predict virality. Some highly shared ones reach over 60 million organic views, which can be more valuable than views on TV. Not only are these voluntary, but they are typically without distraction, and win viewer engagement in the form of likes, comments, or shares. A Super Bowl ad that wins 69 million views on YouTube (e.g., Alexa Mind Reader) costs less than 10 cents per quality view! However, the challenge is triggering virality. We developed a method to predict virality and engineer virality into Ads.
1. Prof. Gerard J. Tellis and co-authors recommended that advertisers use YouTube to tease, test, and tweak (TTT) their ads to maximize sharing and viewing. 2022 saw that maxim put into practice.
2. We developed viral Ads prediction using two scientific models:
a. Prof. Gerard Tellis et al.'s model for viral prediction
b. Deep Learning viral prediction using social media effect
3. The model was able to identify all the top 15 Viral Ads it performed better than the traditional agencies.
4. New proposed method is Tease, Test, Tweak, Target and Spots Ad.
Data Con LA 2022- Embedding medical journeys with machine learning to improve...Data Con LA
Jai Bansal, Senior Manager, Data Science at Aetna
This talk describes an internal data product called Member Embeddings that facilitates modeling of member medical journeys with machine learning.
Medical claims are the key data source we use to understand health journeys at Aetna. Claims are the data artifacts that result from our members' interactions with the healthcare system. Claims contain data like the amount the provider billed, the place of service, and provider specialty. The primary medical information in a claim is represented in codes that indicate the diagnoses, procedures, or drugs for which a member was billed. These codes give us a semi-structured view into the medical reason for each claim and so contain rich information about members' health journeys. However, since the codes themselves are categorical and high-dimensional (10K cardinality), it's challenging to extract insight or predictive power directly from the raw codes on a claim.
To transform claim codes into a more useful format for machine learning, we turned to the concept of embeddings. Word embeddings are widely used in natural language processing to provide numeric vector representations of individual words.
We use a similar approach with our claims data. We treat each claim code as a word or token and use embedding algorithms to learn lower-dimensional vector representations that preserve the original high-dimensional semantic meaning.
This process converts the categorical features into dense numeric representations. In our case, we use sequences of anonymized member claim diagnosis, procedure, and drug codes as training data. We tested a variety of algorithms to learn embeddings for each type of claim code.
We found that the trained embeddings showed relationships between codes that were reasonable from the point of view of subject matter experts. In addition, using the embeddings to predict future healthcare-related events outperformed other basic features, making this tool an easy way to improve predictive model performance and save data scientist time.
Data Con LA 2022 - Data Streaming with KafkaData Con LA
Jie Chen, Manager Advisory, KPMG
Data is the new oil. However, many organizations have fragmented data in siloed line of businesses. In this topic, we will focus on identifying the legacy patterns and their limitations and introducing the new patterns packed by Kafka's core design ideas. The goal is to tirelessly pursue better solutions for organizations to overcome the bottleneck in data pipelines and modernize the digital assets for ready to scale their businesses. In summary, we will walk through three uses cases, recommend Dos and Donts, Take aways for Data Engineers, Data Scientist, Data architect in developing forefront data oriented skills.
DevOps and Testing slides at DASA ConnectKari Kakkonen
My and Rik Marselis slides at 30.5.2024 DASA Connect conference. We discuss about what is testing, then what is agile testing and finally what is Testing in DevOps. Finally we had lovely workshop with the participants trying to find out different ways to think about quality and testing in different parts of the DevOps infinity loop.
Sudheer Mechineni, Head of Application Frameworks, Standard Chartered Bank
Discover how Standard Chartered Bank harnessed the power of Neo4j to transform complex data access challenges into a dynamic, scalable graph database solution. This keynote will cover their journey from initial adoption to deploying a fully automated, enterprise-grade causal cluster, highlighting key strategies for modelling organisational changes and ensuring robust disaster recovery. Learn how these innovations have not only enhanced Standard Chartered Bank’s data infrastructure but also positioned them as pioneers in the banking sector’s adoption of graph technology.
A tale of scale & speed: How the US Navy is enabling software delivery from l...sonjaschweigert1
Rapid and secure feature delivery is a goal across every application team and every branch of the DoD. The Navy’s DevSecOps platform, Party Barge, has achieved:
- Reduction in onboarding time from 5 weeks to 1 day
- Improved developer experience and productivity through actionable findings and reduction of false positives
- Maintenance of superior security standards and inherent policy enforcement with Authorization to Operate (ATO)
Development teams can ship efficiently and ensure applications are cyber ready for Navy Authorizing Officials (AOs). In this webinar, Sigma Defense and Anchore will give attendees a look behind the scenes and demo secure pipeline automation and security artifacts that speed up application ATO and time to production.
We will cover:
- How to remove silos in DevSecOps
- How to build efficient development pipeline roles and component templates
- How to deliver security artifacts that matter for ATO’s (SBOMs, vulnerability reports, and policy evidence)
- How to streamline operations with automated policy checks on container images
Pushing the limits of ePRTC: 100ns holdover for 100 daysAdtran
At WSTS 2024, Alon Stern explored the topic of parametric holdover and explained how recent research findings can be implemented in real-world PNT networks to achieve 100 nanoseconds of accuracy for up to 100 days.
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionAggregage
Join Maher Hanafi, VP of Engineering at Betterworks, in this new session where he'll share a practical framework to transform Gen AI prototypes into impactful products! He'll delve into the complexities of data collection and management, model selection and optimization, and ensuring security, scalability, and responsible use.
Alt. GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using ...James Anderson
Effective Application Security in Software Delivery lifecycle using Deployment Firewall and DBOM
The modern software delivery process (or the CI/CD process) includes many tools, distributed teams, open-source code, and cloud platforms. Constant focus on speed to release software to market, along with the traditional slow and manual security checks has caused gaps in continuous security as an important piece in the software supply chain. Today organizations feel more susceptible to external and internal cyber threats due to the vast attack surface in their applications supply chain and the lack of end-to-end governance and risk management.
The software team must secure its software delivery process to avoid vulnerability and security breaches. This needs to be achieved with existing tool chains and without extensive rework of the delivery processes. This talk will present strategies and techniques for providing visibility into the true risk of the existing vulnerabilities, preventing the introduction of security issues in the software, resolving vulnerabilities in production environments quickly, and capturing the deployment bill of materials (DBOM).
Speakers:
Bob Boule
Robert Boule is a technology enthusiast with PASSION for technology and making things work along with a knack for helping others understand how things work. He comes with around 20 years of solution engineering experience in application security, software continuous delivery, and SaaS platforms. He is known for his dynamic presentations in CI/CD and application security integrated in software delivery lifecycle.
Gopinath Rebala
Gopinath Rebala is the CTO of OpsMx, where he has overall responsibility for the machine learning and data processing architectures for Secure Software Delivery. Gopi also has a strong connection with our customers, leading design and architecture for strategic implementations. Gopi is a frequent speaker and well-known leader in continuous delivery and integrating security into software delivery.
20 Comprehensive Checklist of Designing and Developing a WebsitePixlogix Infotech
Dive into the world of Website Designing and Developing with Pixlogix! Looking to create a stunning online presence? Look no further! Our comprehensive checklist covers everything you need to know to craft a website that stands out. From user-friendly design to seamless functionality, we've got you covered. Don't miss out on this invaluable resource! Check out our checklist now at Pixlogix and start your journey towards a captivating online presence today.
Communications Mining Series - Zero to Hero - Session 1DianaGray10
This session provides introduction to UiPath Communication Mining, importance and platform overview. You will acquire a good understand of the phases in Communication Mining as we go over the platform with you. Topics covered:
• Communication Mining Overview
• Why is it important?
• How can it help today’s business and the benefits
• Phases in Communication Mining
• Demo on Platform overview
• Q/A
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdfPaige Cruz
Monitoring and observability aren’t traditionally found in software curriculums and many of us cobble this knowledge together from whatever vendor or ecosystem we were first introduced to and whatever is a part of your current company’s observability stack.
While the dev and ops silo continues to crumble….many organizations still relegate monitoring & observability as the purview of ops, infra and SRE teams. This is a mistake - achieving a highly observable system requires collaboration up and down the stack.
I, a former op, would like to extend an invitation to all application developers to join the observability party will share these foundational concepts to build on:
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!SOFTTECHHUB
As the digital landscape continually evolves, operating systems play a critical role in shaping user experiences and productivity. The launch of Nitrux Linux 3.5.0 marks a significant milestone, offering a robust alternative to traditional systems such as Windows 11. This article delves into the essence of Nitrux Linux 3.5.0, exploring its unique features, advantages, and how it stands as a compelling choice for both casual users and tech enthusiasts.
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...James Anderson
Effective Application Security in Software Delivery lifecycle using Deployment Firewall and DBOM
The modern software delivery process (or the CI/CD process) includes many tools, distributed teams, open-source code, and cloud platforms. Constant focus on speed to release software to market, along with the traditional slow and manual security checks has caused gaps in continuous security as an important piece in the software supply chain. Today organizations feel more susceptible to external and internal cyber threats due to the vast attack surface in their applications supply chain and the lack of end-to-end governance and risk management.
The software team must secure its software delivery process to avoid vulnerability and security breaches. This needs to be achieved with existing tool chains and without extensive rework of the delivery processes. This talk will present strategies and techniques for providing visibility into the true risk of the existing vulnerabilities, preventing the introduction of security issues in the software, resolving vulnerabilities in production environments quickly, and capturing the deployment bill of materials (DBOM).
Speakers:
Bob Boule
Robert Boule is a technology enthusiast with PASSION for technology and making things work along with a knack for helping others understand how things work. He comes with around 20 years of solution engineering experience in application security, software continuous delivery, and SaaS platforms. He is known for his dynamic presentations in CI/CD and application security integrated in software delivery lifecycle.
Gopinath Rebala
Gopinath Rebala is the CTO of OpsMx, where he has overall responsibility for the machine learning and data processing architectures for Secure Software Delivery. Gopi also has a strong connection with our customers, leading design and architecture for strategic implementations. Gopi is a frequent speaker and well-known leader in continuous delivery and integrating security into software delivery.
Removing Uninteresting Bytes in Software FuzzingAftab Hussain
Imagine a world where software fuzzing, the process of mutating bytes in test seeds to uncover hidden and erroneous program behaviors, becomes faster and more effective. A lot depends on the initial seeds, which can significantly dictate the trajectory of a fuzzing campaign, particularly in terms of how long it takes to uncover interesting behaviour in your code. We introduce DIAR, a technique designed to speedup fuzzing campaigns by pinpointing and eliminating those uninteresting bytes in the seeds. Picture this: instead of wasting valuable resources on meaningless mutations in large, bloated seeds, DIAR removes the unnecessary bytes, streamlining the entire process.
In this work, we equipped AFL, a popular fuzzer, with DIAR and examined two critical Linux libraries -- Libxml's xmllint, a tool for parsing xml documents, and Binutil's readelf, an essential debugging and security analysis command-line tool used to display detailed information about ELF (Executable and Linkable Format). Our preliminary results show that AFL+DIAR does not only discover new paths more quickly but also achieves higher coverage overall. This work thus showcases how starting with lean and optimized seeds can lead to faster, more comprehensive fuzzing campaigns -- and DIAR helps you find such seeds.
- These are slides of the talk given at IEEE International Conference on Software Testing Verification and Validation Workshop, ICSTW 2022.
UiPath Test Automation using UiPath Test Suite series, part 5DianaGray10
Welcome to UiPath Test Automation using UiPath Test Suite series part 5. In this session, we will cover CI/CD with devops.
Topics covered:
CI/CD with in UiPath
End-to-end overview of CI/CD pipeline with Azure devops
Speaker:
Lyndsey Byblow, Test Suite Sales Engineer @ UiPath, Inc.
2. Who am I?
2
• I'm Kyle Polich
• I work at DataScience
• I hostThe Data Skeptic Podcast
• I’m excited to share some ideas about data
mining framed around the Apriori Algorithm
• And examples on open data you can
reproduce
3. Outline
3
• What is Association Mining?
• The Apriori Algorithm
• Examples
• Big Data
• Criticisms
• Tips andTricks
4. General Concept
4
• Unsupervised Learning
• Association rule learning (A and B) (A and B and C)
• If N items, than 2N-1 itemsets (powerset w/o empty)
• Common itemsets are made up of common
sub-itemsets
• Iteratively build candidates based on frequency
6. Isn’t this a dead algorithm?
6
Well, the apriori algorithm might be outdated
but a) this page is about that algorithm! and
b) not necessary to state,
but it is the first significant algorithm, and
the basic idea is used again and again in
several succeeding algorithms
so it is important to understand it.Exa 18:33,
16 May 2007 (UTC)
Excerpt fromWikipedia talk page
By user 81.104.165.184
8. Isn’t this a dead algorithm?
8
C4.5
Apriori algorithm
Hyperloglog
9. Isn’t this a dead algorithm?
9
Google Scholar tracks 18,286
citations
TODO: visualize this as a time series
10. Isn’t this a dead algorithm?
10
1. Easy to learn in a 30 minute session
2. Always start simple, and grow in complexity
3. Simple, but still powerful
4. Practical to implement
5. Runs well at scale
6. Good study of algorithmic design
7. I believe it’s a useful algorithm
11. Origin / Creators
11
Fast Algorithms for Mining Association Rules
Rakesh Agrawal & Ramakrishnan Srikant
IBMAlmaden Research Center
20th InternationalConference onVery Large Data Bases
Santiago, Chile - September 1994
http://rakesh.agrawal-family.com/papers/vldb94apriori.pdf
12. Key Concept: Associative Rules
12
• “Peanut Butter” AND “Jelly”
• “Sausage”AND “mustard” AND “deli roll”
• “Good schools” AND “easy parking” AND
“walk to restaurants”
22. Metrics
22
Support
% of cases containing itemset
R and Machine Learning (5)
Benjamin Uminsky
Gian Gonzanga
Jim Mcguire
Kyle Polich
Szilard Pafka
Everyone (35)
Aaron Wepler, Abhi Nemani, Adam Mollenkopf, Alan Gates, Amelia
Mcnamara, Arvind Prabhakar, Ashish Singh, Benjamin Uminsky, Bikas Saha,
Brian Kursar, Chris Fregly, Felix Chern, Gian GonzangatH, Hyunsik Choi, Jeff
Morris, Jim Mcguire, John De Goes, Jonathan Gray, Josiah Carlson, Karen
Lopez, Khanderao Kand, Kyle Polich, Michael Limcaco, Michael Stack,
Rachel Pedreschi, Raj Babu, Romain Rigaux, Sabri Sansoy, Szilard Pafka,Tim
Ellis,Tim Fulmer, Ulas Bardak,Vinayak Borkar, Will Ochandarena, ZainAsgar
5 / 35 = .14286
23. Metrics
23
Confidence
% of cases containing itemset
R (6)
Amelia Mcnamara, Benjamin Uminsky, Gian Gonzanga, Jim
Mcguire, Kyle Polich, Szilard Pafka
Machine Learning (7)
Benjamin Uminsky, Brian Kursar, Gian Gonzanga, Jim
Mcguire, Kyle Polich, Szilard Pafka, Ulas Bardak
R -> Machine Learning
5 / 7 = .71286
24. CodeWalkthrough
24
Let minimum support = .19
name count support
Algorithms 7 0.2
Machine Learning 7 0.2
Software Engineering 7 0.2
Software Development 9 0.257143
Distributed Systems 11 0.314286
Java 12 0.342857
Big Data 13 0.371429
Hadoop 14 0.4
25. CodeWalkthrough
25
Let minimum support = .19; k=2
name count support
Algorithms 7 0.2
Machine Learning 7 0.2
Software Engineering 7 0.2
Software Development 9 0.257143
Distributed Systems 11 0.314286
Java 12 0.342857
Big Data 13 0.371429
Hadoop 14 0.4
26. CodeWalkthrough
26
Let minimum support = .19; k=2
name count support
Algorithms 7 0.2
Machine Learning 7 0.2
Software Engineering 7 0.2
Software Development 9 0.257143
Distributed Systems 11 0.314286
Java 12 0.342857
Big Data 13 0.371429
Hadoop 14 0.4
Algorithms Hadoop
Software
Development Distributed Systems
Hadoop
Distributed
Systems Big Data Distributed Systems
Java Hadoop
Software
Engineering Distributed Systems
Software
Development Hadoop Distributed Systems Machine Learning
Hadoop Big Data
Software
Development Java
Hadoop
Software
Engineering Java Big Data
Hadoop
Machine
Learning Java Software Engineering
Algorithms
Distributed
Systems Java Machine Learning
Java Algorithms
Software
Development Big Data
Software
Development Algorithms
Software
Development Software Engineering
Algorithms Big Data
Software
Development Machine Learning
Algorithms
Software
Engineering
Software
Engineering Big Data
Algorithms
Machine
Learning Big Data Machine Learning
Java
Distributed
Systems
Software
Engineering Machine Learning
27. CodeWalkthrough
27
Let minimum support = .19; k=2
name count support
Algorithms 7 0.2
Machine Learning 7 0.2
Software Engineering 7 0.2
Software Development 9 0.257143
Distributed Systems 11 0.314286
Java 12 0.342857
Big Data 13 0.371429
Hadoop 14 0.4
Algorithms Hadoop 3
Software
Development Distributed Systems 4
Hadoop
Distributed
Systems 10 Big Data Distributed Systems 7
Java Hadoop 8
Software
Engineering Distributed Systems 3
Software
Development Hadoop 4 Distributed Systems Machine Learning 0
Hadoop Big Data 8
Software
Development Java 4
Hadoop
Software
Engineering 2 Java Big Data 5
Hadoop
Machine
Learning 1 Java Software Engineering 3
Algorithms
Distributed
Systems 4 Java Machine Learning 1
Java Algorithms 4
Software
Development Big Data 4
Software
Development Algorithms 3
Software
Development Software Engineering 5
Algorithms Big Data 2
Software
Development Machine Learning 0
Algorithms
Software
Engineering 3
Software
Engineering Big Data 2
Algorithms
Machine
Learning 2 Big Data Machine Learning 2
Java
Distributed
Systems 8
Software
Engineering Machine Learning 0
28. CodeWalkthrough
28
Let minimum support = .19; k=2
name count support
Algorithms 7 0.2
Machine Learning 7 0.2
Software Engineering 7 0.2
Software Development 9 0.257143
Distributed Systems 11 0.314286
Java 12 0.342857
Big Data 13 0.371429
Hadoop 14 0.4
Algorithms Hadoop 3
Software
Development Distributed Systems 4
Hadoop
Distributed
Systems 10 Big Data Distributed Systems 7
Java Hadoop 8
Software
Engineering Distributed Systems 3
Software
Development Hadoop 4 Distributed Systems Machine Learning 0
Hadoop Big Data 8
Software
Development Java 4
Hadoop
Software
Engineering 2 Java Big Data 5
Hadoop
Machine
Learning 1 Java Software Engineering 3
Algorithms
Distributed
Systems 4 Java Machine Learning 1
Java Algorithms 4
Software
Development Big Data 4
Software
Development Algorithms 3
Software
Development Software Engineering 5
Algorithms Big Data 2
Software
Development Machine Learning 0
Algorithms
Software
Engineering 3
Software
Engineering Big Data 2
Algorithms
Machine
Learning 2 Big Data Machine Learning 2
Java
Distributed
Systems 8
Software
Engineering Machine Learning 0
29. CodeWalkthrough
29
Let minimum support = .19; k=3
name count support
Hadoop, Distributed Systems 10 .35
Java, Hadoop 8 0.22857
Hadoop, Big Data 8 0.22857
Java, Distributed Systems 8 0.22857
Big Data, Distributed Systems 7 0.2
Hadoop Distributed Systems Java 7 0.2
Hadoop Distributed Systems Big Data 7 0.2
30. CodeWalkthrough
30
Let minimum support = .19; k=3
name count support
Hadoop, Distributed Systems, Java 7 0.2
Hadoop, Distributed Systems, Big Data 7 0.2
Hadoop
Distributed Systems
Java
Big Data
1. Alan Gates
2. Ashish Singh
3. Jonathan Gray
4. Michael Stack
5. Vinayak Borkar
31. CodeWalkthrough
31
Let minimum support = .19; k=4
Hadoop
Distributed Systems
Java
Big Data
1. Alan Gates
2. Ashish Singh
3. Jonathan Gray
4. Michael Stack
5. Vinayak Borkar
33. Computational Commentary
33
• Outer loop should
(presumably) be a small
number of iterations
• Be careful selecting your
minimum!
• Maybe put a max iterations?
35. Computational Commentary
35
• This can be the “map” step
• Pseudo code a bit unclear
here
• Could be highly optimized
• Can run in O(n) time with
pre-built hash tables
40. Recipes - Single Itemsets
40
garlic onion parsley
all purpose flour salt vanilla extract
canola oil chicken broth onion
all-purpose flour almond extract brown sugar
baking powder butter softened cinnamon
all-purpose flour baking powder sugar
brown sugar milk sugar
cilantro olive oil red onion
all purpose flour butter softened sugar
bay leaves oregano parmesan cheese
ginger soba noodles toasted pine nuts
41. Los Angeles 311 Data
41
Blocked Driveways Bulky Item Pick-up
Holiday Trash Collection Internal Affairs Group - LAPD
Report Broken Parking Meters Abandoned Vehicles
Complaint - LAPD (How to Make
a Complaint) Bulky Item Pick-up
Animal Service Centers Report streetlight outages
Police Auctions Blocked Driveways
Sprinklers Running at Parks Bulky Item Pick-up
Graffiti Removal - Community
Beautification
877 ASK-LAPD - Non-emergency
Police Service
LADWP Central Operator Constituent Service Office of the Mayor
42. Frequent itemset mining in games
42
• Anders Drachen has written about Apriori applications in gaming
• http://bit.ly/1Fi8vHu
45. Online Feature Discovery in
Relational Reinforcement Learning (2006)
45
Presented at the ICML Workshop on Open Problems in Statistical Relational Learning,
Pittsburgh, PA, 2006
Scott Sanner, University ofToronto
• Reinforcement learning
• Used to identify for focusing on frequently visited areas of the state
space when doing structure learning
46. A Novel Modified Apriori Approach for
Web Document Clustering (2015)
46
Computational Intelligence in Data Mining-Volume 3, 159-171, 2015
Roul,Varshneya, Kalra, Sahay
• Keywords / ngrams as items; documents as itemsets
• Centroid describes topic / theme of pages
• Decrease candidate itemsets during candidate generation
• Only consider itemsets in a specific iteration
• Some code optimizations around unnecessary steps
51. Repeated database table scans
51
• Distributed solutions can solve this on large
datasets
• In-memory analysis can solve for small
52. Fails to observe rare but important matches
52
• Described as “weak” associative rules
• Example fromThe Elements of Statistical
Learning by Hastie,Tibshirani, and Friedman
is “caviar” and “wine”
• Adaptations of the algorithm could address
this
55. Great for Ensembling
55
• Quick and dirty unsupervised analysis
• Get initial glimpse into a new dataset
• Feed results into other approaches
56. Optimize forYour Use Case
56
• TODO: Hive trick
• Find efficient data structure to capture your
transactions
57. Market Basket / Affinity Analysis
57
Purpose
• Identify cross-selling / up-selling opportunities
• Shelf / aisle placement optimization
The Apriori Algorithm…
• provides an easy, fast, first look
• is useful in creating a feature label variable
called “has common itemset”
• turns out great results in ensemble
approaches
58. 58
The Apriori Algorithm is worth your time.
• Informative when studied
• Unsupervised, great starting point
• Extendable
• Great as an ensemble approach
CONCLUSION
Google Trend shows reasonable interest, even today
Holding better than C4.5, more interesting than hyperloglog
2 – point in right direction
6 – we need to study more, digital red lining
I will go step by step through this, the subtleties are important
Gets all potential itemsets based on the previous iteration. Assume itemsets made up of common item subsets
Originally database. I use in-memory hash tables
Very expensive looping over T – database scan
Pulled speakers skills from linkedin
R and Machine Learning
Initialize all 1 element datasets – too many to show here, set .19 as support parameter
Set k=2, check L1, start
Apriori-gen step generates all possible rules based on the previous rules. Given what is in upper right, all pairs
Here are all the counts
Filter out those below our minimum sensitivity
Do the next iteration of k
Only 5 people have the available combination of popular skills. Not enough for minimum support…
Thus, loop is done
Our final results
Few iterations
t \in T is a database call in the original iteration; fine because you should have a small number of iterations
I pre-calculate a hash table mapping 1-itemsets to a hash of the transactions that contain it
Thus n = k
I pre-calculate a hash table mapping 1-itemsets to a hash of the transactions that contain it
Thus n = k
Trade off, not smooth because small data
You’ll notice my dataset isn’t perfectly clean. I could have cleaned more, but I like to leave some dirt to measure the resilience and to measure the iterative improvement.
You’ll notice my dataset isn’t perfectly clean. I could have cleaned more, but I like to leave some dirt to measure the resilience and to measure the iterative improvement.
Also, some of these are interesting, some are not.
Comment on their work with only one trip to the database
Also, Tristan’s suggestion
Also, Tristan’s suggestion
Also, Tristan’s suggestion
Most baskets are lognormal – how do you get to the interesting stuff? Focus on ensembling